Devcon #6 - Bartek Kiepuszewski

Welcome, Bankless Nation, back to the DevCon 6 experience. I hope you've been enjoying these interviews thus far. I hope they've been making you feel like you were on the ground at DEF CON. It was a treat. In this interview, we're talking with Bartek. Bartek is the leader, the co-founder of Layer 2 Beat. He's previously Maker DAO, and he is at the middle of the Layer 2 wars. Layer 2 Beat is this place where all of these layer twos are like indexed and documented about what holes they have, what they need to work on. And it's been like the sparring ground for all of these layer twos. So I get Bartek's perspective as to what it's like to be in the middle of the layer two wars and how he thinks the layer two wars are progressing from here. And overall, are we as Ethereum community members not critical enough about our own layer twos? Are we hypocrites for critiquing cross layer one bridges, other alt layer ones, while also not looking at our own layer twos and seeing the holes and flaws in them as well? I asked him that question and more all about the layer two wars, the very, very spicy layer two wars. So I hope you enjoyed this conversation with Bartek from Layer Two B.

Bank of this nation, I'm here with Bartek, founder of Layer 2B and also a contributor at MakerDAO. And Bartek has found himself, I think, at the center of these Layer 2 wars, I wish there have been a ton of announcements here at DevCon Bogota. Bartek, thank you so much for coming on and joining me here.

It's my pleasure.

I've been drawing Columbia.

Well, so far so good. Yeah, I like the country. You know, it's the first time that I'm in Colombia, like probably most people at DEF CON. And DEF CON is always a great opportunity to see other places and cultures. And I really appreciate the approach of the Human Foundation where I feel like we're trying to include everybody, right? It's not like just Europe and US and Europe and US, like most other conferences. But we actually been to Japan, you know, we are here now.

I'm truly wondering, you know, what's next.

Mm-hmm. Well, certainly what's next. There's so many things that's next in Ethereum, and Layer 2 is definitely one of those things. Like I said, you are the co-founder of Layer Two Beat, which is this website that we go to to look at how many holes there are in the layer two ecosystem. And Vitalik recently at the roll-up day on Monday was giving a presentation as to all these different ways that we need to fix layer twos. How do you think that progress is going? And are there any announcements that you've seen here at DevCon Bogota that is working in that direction?

Well, frankly, for uh so many people that I talk to that, you know, I don't like attend a lot of sessions. I haven't seen a lot of announcements. Uh having said that, I think that most teams they have presented roadmaps uh which are fully decentralized. You know, they promise all of us to shut the training wheels. And I think if we were to listen to uh list on L2P just pure rollups, we probably wouldn't list literally anyone, right? It would be a very boring website. So right now, you know, as it stands, every single project is very much centralized and is very much controlled by the team that actually created the rollups. But to my knowledge, with almost no exception, each one of them has a roadmap, like a decentralization roadmap. And I think right now, you know, we're

trying to kind of track the progress if you like, right? And made them transparent and made them honest. And I think we're getting there, for sure.

And one of the critiques that we've seen outside of the Ethereum space and sometimes also inside of the Ethereum space is that we're promoting these layer twos as if they're the magical secure solution to scale a chain. But then we're also kind of hiding and pushing under the rug the fact that these have multi sig bridge contracts and we're missing fraud proofs. Do you think that the Ethereum community or the like Ethereum promoters, maybe me and Ryan are guilty of this, are like hypocritical of the way that we kind of pitch layer twos? Or do you think there's a more nuance here?

Well uh the this is a very loaded question.

So I'd say that you know every single team they've got their own agenda and some of them are more, let's say, transparent than the others. Some of them are hiding behind the marketing and whatnot. Everyone seems to be extremely keen at building the community. And well, we cannot hide the fact that it is a race, right? I wouldn't call it a war, and we don't know exactly how the future of layer two ecosystem on Ethereum will look like. Are we gonna see many different rollups or maybe one will take it all? So that's still an open question in my mind. But if one team sees the other team like launching something which may have a lot of training wheels, you know, they might actually be pressed to do the same. And the general community, I think, they don't see all these nuances because they may not be published widely, right? Sometimes we're not published at all. And this is why you know we essentially created this website so that we disclose some of these things. Uh, and we just want to make all these teams honest and we want to make sure that the users actually understand all the security assumptions behind all these systems.

You've talked about the roadmap that many of these layer two teams have presented to solve and fix some of these holes. How credible are these roadmaps? Are they legit roadmaps? There's real tech here, or is it more just hand waving?

So that would probably differ. Every single team has a little bit different approach. So if we look at the optimistic rollups, for example, we see two major teams putting a lot of work into actually making sure that all these fraudproof technology is working and EVM equivalent systems will eventually be deployed. So I'm talking about Arbitrum and Optimism, and we've got some other teams that are forking mainly optimism code, and they're kind of waiting for optimism to deliver the fraud proof system. So you can see that you know some teams are innovating, and some teams are trying to sort of you know fork the code and maybe try to innovate on the edges. I'm mainly thinking about the Boba and Mattis. So that's optimistic for the ZK. Most of the teams are innovating like crazy. I mean, this is like amazing, really, to see the progress. I think we've all been surprised about all these teams and well, Starknet, ZK Sync, Aztec, right? Scroll, I mean, it's just unbelievable polygon as well. Everyone is like pushing code really hard, and every single conference that I go to, there's like new announcements, right? From either one of them. So

My understanding is that ZK Sync will launch very soon. Starknet is just upgrading to version one Cairo, which is like the major, major upgrade. The first version was very much centralized and wasn't particularly censorship resistant. We've pushed them really, really hard, you know, to work on this problem. And I believe that they listen to the community, and the new version, which is like way more complicated, will actually allow people to force transactions to their system, right? So they cannot be censored.

Uh-huh. I remember d when I was reporting on ETHCC back in April of earlier this year, whenever that was. I can't remember how many months ago.

Maybe

June something. Yeah. Jul

whatever.

July. Yeah.

July.

Right, right.

It was July.

People are joking that the theme of the ECC was the ZK EVM because so many people announced their ZK EVMs. So it was a test net or launch date or something. That also seems to be the theme of the rollups this week as well, here at DEF CON Bogota, where Polygon released their ZK EVM test net, Scroll released their ZK EVM test net, ZK Sync released their layer three test net, and they have their mainnet in just like 10, 20 days here. And also something that Vitalik said in his presentation at Rollup Day was the ZK teams seem to be like the one part of like crypto building that has been massively accelerated beyond expectations. Well, you know, Ethereum, the merge was slow, sharding's been slow, but the ZK EVMs are here. When all of these mainnets do come, like

first off, how big of a deal is that? And like, are they going to come and is it going to be as a big of a thing that people are excited for? Or is it still going to be like, okay, main net, but users are gonna have to wait another like, I don't know, six months to be able to do something? Like, overall, what's your take on the trajectory of the ZK ecosystem?

Well, I mean right now we have to be able to sort of differentiate between systems that are ZK equivalent or EVM equivalent or and EVM compatible. So ZK Sync is a good example of you know ZK rollup that is compatible with V VM, but this is not like equivalent system, right? And I think we're gonna see a very interesting competition between systems that try to emulate EVM as close as possible and systems that will prefer to make sure that the proving is like optimized for the particular VM, right? And in this category, I would put definitely ZK Sync and Starknet. So the competition between these two approaches would be very interesting to watch. Having said that, I think end users probably for them, you know, it wouldn't really matter. It might make a small difference for developers. But my thinking about this is that it doesn't matter so much, you know, how friendly this is for developers. I mean, it's important that you know it's easy to port your code, it might be important to make sure that your code is secure. But for end users, it's much more important that the system is fast, it's cheap, right? Uh it's got the big ecosystem for good wallet support. And I think for uh to that end, you know, the account abstraction is like super important. And uh they won't see much of a difference, frankly, right? Even today, when you're using you know Starknet or Optimism or Arbitrum, I mean, you won't notice much of a difference. So we're like debating the details that most end users won't even notice, right? And you know, it will come to the edges, I think, if something goes really wrong. Like uh if one of these systems will start to censor.

Mm-hmm.

Then end users will learn that there is a difference, right? And then they will see that there are some systems that you know have working escape hatches, whilst some other systems don't have that facility and suddenly your phones are stuck. Right. Because the you know rel up operators decided to be, I don't know, off compliant or whatever, right? Pressed by the regulators. So, you know, like for the daily usage, uh, I don't think the users will actually notice much. Uh, but I think it's our responsibility to educate them. What are those security assumptions, right?

Do you have your own personal opinion on the debate between ZK EVM equivalence versus EVM compatibility? Or are you saying both sides have strong arguments, let the best roll-up win?

Uh I think both sides presented very strong arguments. You know, I had the pleasure of moderating the panel on the roll-up day on that very topic, you know, and uh and I can see passionate, you know, statements from both sides. So it will be extremely interesting to watch. And long term, maybe going ZKM equivalent, you know, makes sense. But you know, short term, I think are certain advantages right now to have a VM that's like ZK friendly because it will allow you to do things faster and cheaper, and time to market is likely to be quicker, right?

All right. One of the things that has been living rent-free in my brain, and the thing I've been trying to learn about the most here at DevCon is Optimism's Bedrock. And it's interesting to see the modularity of Ethereum find its way also into the layer twos. And that's really what Optimism Bedrock is, is a modular layer twos. In the same way that we have a modular layer one Ethereum. My mind here is that like what Optimism's Bedrock is doing in the OP stack is that they're reducing the barrier to forking and producing a new layer two. And as somebody who has to manage a website with that is a registry of all the properties of all these layer twos, I'm slightly worried for you, Bartek, because what happens if you see a Cambrian explosion of optimistic roll ups, layer threes, do you see the same future as well? Or how do you think about managing all these proliferation of layer twos that might

I mean, most definitely, and even right now, the website is hardly manageable, and I'm sure that there are certain inaccuracies and mistakes. And in future months, you know, the most of the engineering effort will go into automation. I mean, at the end of the day, everything ends up on chain. And so we uh with our monitoring infrastructure, once you know it will be developed, we will actually get all the information automatically from chain, right? There must be some registry of all these rollups. There will be a way to sort of read their configuration directly from blockchain. So hopefully, if they follow the same cookie-cutter approach and like fog and create their own rollup, it will be just a matter of you know having our bots to read the data and fill in the website. And then it will be a question of how to actually present it to user in a manageable way, right? Which is, I guess, you know, a question for the UI designers. So I'd say that today it used to be 80% of the effort for the research and 20% of the effort for you know, like presenting all the information, but now it's gonna be reverse almost, right? Um the automation infrastructure behind L2 beats is gonna be actually quite complex. And um tracking all the changes will be possible because you know, finally we've got some fantastic tooling like the token flow insight. Uh, this is the analytical database that tracks every single state on Ethereum in all Ethereum history, right? So we can just watch particular slots and it will inform us you know what exactly happened and when it happened, and uh it will trigger warnings, essentially, right? So that's how I kind of think about it. I mean, it it has to be managed by essentially automation.

Sure. And of course, layer twos under the hood are extremely complicated. I don't even know how to code. So when something goes down, I wouldn't know what to do. And so, Bartek, what you're doing at Layer 2B, in my mind, is just producing user sovereignty and tools that the individuals can use that are non technical and very, very helpful. So thank you for everything you're doing at Layer 2B. And I hope you have a fantastic rest of your DEF CON.

Okay, thank you very much.

Cheers.